Deblurring for non-2D Fourier transform magnetic resonance imaging

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging (MRI)

Background:

• Off-resonant reconstruction in non-2D Fourier transform imaging causes blurring and point spread function changes.
• This blurring is a known issue in projection reconstruction and spiral k-space scanning sequences.
• Existing methods often require resonant frequency maps to correct for these artifacts.

Purpose of the Study:

• To introduce a novel method for automatic blur removal in non-2D Fourier transform imaging.
• To improve the utility of projection reconstruction and spiral k-space scanning sequences.
• To eliminate the need for resonant frequency maps in blur correction.

Main Methods:

• Modulating raw data to different frequencies and reconstructing base images.
• Calculating a focusing measure for each point in each base image to determine blur.
• Constructing a composite image using unblurred regions from the base images.

Main Results:

• Successfully applied the method to phantom and in vivo images.
• Demonstrated effective removal of blur caused by magnetic field inhomogeneity and susceptibility.
• Validated the method's performance in projection-reconstruction and spiral-scan sequences.

Conclusions:

• The developed method effectively removes blur without requiring a resonant frequency map.
• This technique significantly enhances the quality and usefulness of non-2D Fourier transform imaging methods.
• The approach shows promise for improving MRI applications in both research and clinical settings.